- Hussein Abbass, DSARC and UNSW@ADFA, Australia
- Capt David Dunwoody, 14 Software Engineering Squadron, 14 Wing Greenwood, Canadian Forces, Canada
- Leonid Perlovsky, Air Force Research Laboratory and Harvard University, USA
- Dale Reding, DRDC Centre for Operational Research and Analysis, Canada
Invited Tutorial Speakers
Kathleen Carley, Institute for Software Research at Carnegie Mellon University, USA
Dipankar Dasgupta, Center of Information Assurance at the University of Memphis, USA
Simon Haykin, Cognitive Systems Laboratory, McMaster University, Canada
Forward, Reverse and Emerging Dynamics: Can Complex Adaptive Systems Play a Game with the Unknown?
Prof. Hussein Abbass is the Chair of Information Technology at the School of Information Technology and Electrical Engineering, University of New South Wales, the Australian Defence Force Academy in Canberra, Australia. He is the Director of the University Defence and Security Applications Research Centre and the Director of the Artificial Life and Adaptive Robotics Laboratory. He is a fellow of the Australian Computer Society (ACS), an Associate Fellow of the Australian Institute of Management (AIM), a senior member of the IEEE, the chair of ACS National Committee on Complex Systems, the chair of the IEEE Task Force on Complex Adaptive Systems and Artificial Life, and a member of a number of national and international committees including the IEEE technical committee on Data Mining and the IEEE working group on soft computing in the SMC society. He holds an Advisory Professor at Vietnam National University, Ho-Chi Minh city, and held visiting positions at Imperial College London and University of Illinois. He published more than 180 papers, and his main research interests include evolutionary games, learning (data mining) and optimization, ensemble learning, and multi-agent systems. He has 180+ refereed publications and his research is funded by the Australian Research Council (ARC), Eurocontrol, and other government organisations and industry.
Human Centred Automation for UAVs: A Bayesian Network Approach
This presentation discusses the use of computer intelligence as applied to human centred automation (HCA) to increase operator-uninhabited airborne vehicle (UAV) interaction. This presentation will examine three areas of research as they apply to this design methodology. The first area examines HCA and its use to overcome the risks of automation. HCA states that the automated system must be able to monitor the operator. To do so, the system must first model the goals and tasks of the operator's interaction with the UAV. An Operator Function Model (OFM) design methodology models those tasks carried out by a human operator to achieve a specified goal. Finally, the UAV uses a Bayesian networking to decide what tasks and, ultimately the goal, the operator is carrying out. From this point, the UAV can now monitor the operator and aid them in completing their mission. This presentation recommends the application of HCA to all future developments of UAV automation.
Capt David Dunwoody originates from Southern Manitoba. In 1991 he enrolled in 402 Sqn as an Air Reservist. After completing General Military Training, Capt Dunwoody trained at the Canadian Forces School of Communication and Electronics. He earned his qualifications as an Instrument Electrician in 1992 at CFB Borden. In 1998 Capt Dunwoody was accepted into the Regular Force. He earned his Bachelor's of Art in October 2001, while studying to become an Air Navigator. He majored in psychology with a minor in computer science. He graduated from the Canadian Forces Air Navigation School in 2002. Subsequently he was posted to 415 Sqn at 14 Wing Greenwood, Nova Scotia to serve on the CP-140 Aurora. Prior to commencing his aircrew training on the Aurora he was attach posted to 5 Wing Goose Bay in the position of Duty Operations Officers. In February 2003, he began his training as an Acoustic Sensor Operator graduating from the course in June of that year. In 2004, Capt Dunwoody volunteered to serve as the 2 I/C Air Tasking authority for Op Athena, Roto 2 in Kabul, Afghanistan. This mission was part of the UN International Security Assistance Force. He served from August 2004 to February 2005. Upon return Capt Dunwoody resumed his duties with 415 Sqn. 415 Sqn was closed down in June 2005 and Capt Dunwoody was transferred to 405 Sqn, also located at 14 Wing Greenwood. In July 2007 he was transferred to 17 Wing Winnipeg to attend the Aerospace Systems Course at the CF School of Aerospace Studies. After graduating from the Aerospace Systems Course in 2008, Capt Dunwoody was posted to 14 Software Engineering Squadron at 14 Wing Greenwood. He currently serves as the Operational Liaison and Acceptance Officer as well as project officer for several projects.
Detecting Weak Signals and Predicting Cultures
Detecting weak patterns in clutter is difficult: parameters of pattern models have to be estimated jointly with association-assignment that is deciding which signals come from the object and which from clutter. Association problem leads to combinatorial complexity, and performance is bounded by the number of operations, not by information in signals. Algorithms under-perform information-theoretic limits by orders of magnitude. The talk describes dynamic logic (DL), which resulted in orders of magnitude improvement of detection performance in signal-to-clutter ratio.
DL is a process-logic from vague-to-crisp. It evolves vague-fuzzy initial states into crisp-logical final states of decision-making. This process-logic made possible the mathematical breakthrough. DL is inspired by mechanisms of the mind. It models the mind from lower-level perception to higher-level cognition including beautiful and sublime, to interaction between cognition and language. Recent neuroimaging experiments proved that DL is used in perception and supported DL theory of cognition- language interaction.
Cultures are made of people, so from modeling the minds DL has been further developed to model cultures. The developed theory is related to Sapir-Whorf hypothesis (SWH): languages may affect thinking and culture. Similar ideas were previously suggested by Bhartrihari (India 6th c), Humboldt and Nietzsche (19th c.). The DL theory corresponds to a new, emotional version of this old hypothesis: emotional differences among languages are no less important than semantic ones. Three types of solutions have been obtained, corresponding to three types of cultures. The low-emotional knowledge-acquiring cultures lead to science and technology, but also to internal instabilities related to doubts in knowledge and values. The high-emotional traditional cultures are internally stable, the knowledge and values are so highly esteemed that no deliberation is possible and knowledge stagnates. Multi-language cultures with vibrant cultural exchanges could accumulate knowledge while remaining stable.
Future research include: measure parameters of culture-evolving equations; multi-agent culture simulations; language-understanding search engines; integrating multi-sensor information; multi-agent simulations integrating language and cognition; the role of music in cognition and culture; account for music in culture evolution.
Dr. Leonid Perlovsky, Principal Research Physicist and Technical Advisor at the Air Force Research Lab, Visiting Scholar at Harvard. Leads programs on sensor detection and fusion, cognitive algorithms, socio-cultural modeling. From 1985 to 1999, as Chief Scientist at Nichols Research, a $0.5B high-tech organization, led the corporate research in intelligent systems, neural networks, and sensor fusion. He served as professor at Novosibirsk and New York Universities; participated as a principal in startups developing tools for text understanding, biotechnology, and financial predictions. His company predicted the market crash following 9/11 a week before the event, detecting ripples of illicit Al Qaeda trades, and later helped SEC looking for perpetrators. He delivered invited keynote plenary talks and tutorial lectures around the globe, published more then 320 papers, 10 book chapters, and 3 books including a monograph Neural Networks and Intellect, Oxford University Press, 2001 (currently in the 3rd printing). Dr. Perlovsky organizes conferences on Computational Intelligence, leads IEEE NNTC Task Force on The Mind and Brain, serves on the Board of Governors of the International Neural Network Society 2008-2011, as Associate Editor for IEEE Transactions for Neural Networks, Editor-at-Large for Natural Computations, and Editor-in-Chief for Physics of Life Reviews, which he started jointly with Nobel laureate I. Prigogine. He received several National and International Awards, including Gabor Award, the highest engineering award from International Neural Network Society 2007; and McLucas Award from the USAF 2007 (the highest AF scientific award).
Machine Learning Methods for Significant Incident Characterization
Dale Reding is Chief Scientist, Defence Research and Development Canada - Centre for Operational Research and Analysis (DRDC CORA).
Mr. Reding holds a Masters degree in theoretical Physics with over 20 years experience in the application and development of operational research and analysis methods, across a broad spectrum of defence problems. He has served as a Canadian defence scientist on the operational research staffs within National Defence Headquarters (Logistics, Land, Maritime), NORAD/US Space Command (Colorado Springs) and at the NATO C3 Agency (The Hague). In recognition of his work at NORAD, he was awarded the NORAD Deputy Commander in Chief Commendation. As a scientific manager at National Defence Headquarters, Mr. Reding was the Director - Operational Research (General Analysis), responsible for analytical support in the areas of logistics, personnel, strategic analysis and methodology development. In his current role as Chief Scientist, Mr. Reding leads the advancement of scientific capability and professional development within DRDC CORA.
Mr. Reding's research interests lie in the areas of inverse theory, computational intelligence, semantic analysis, modeling & simulation and the study of complex systems.
Invited Tutorial Speakers
Dynamic Network Analysis and Security
Dynamic Network Analysis (DNA) is the study of how entities are
constrained and enabled by the relations among them and the process that
lead to change in these relations. Methodologically, it draws on
social networks analysis, link analysis, multi-agent modeling, machine
learning, graph theory, and non-parametric statistics to assess complex
meta-networks. A meta-network is a multi-mode (many types of nodes),
multi-plex (many types of links), multi-level (many types of networks
among the same classes of nodes) system in which both the nodes and the
links have attributes, some of which can change with time.
This session provides an overview of DNA, describes the key tools of
relevance, describes recent advances, reviews applications in the security and defense arena, and discusses challenges for work in this
area. Core technologies include techniques for collecting DNA
information (e.g., text-mining with AutoMap, assessment and
visualization of the meta-network (e.g. using ORA), and examining
potential changes and their impact using multi-agent simulation (e.g.,
using Construct). DNA has a long history of utility in the security area; and indeed has been used to identify connections among gangs, identify key actors and groups who are impacting social outcomes, and to assess various courses of action for disrupting groups. Recent advances in linking spatial data and social network data, and in examining change in networks over time, have further enhanced the value of DNA for security. For example, it is now possible to identify local spatio-temporal patterns and shifts in activity related to certain
outcomes of interest.
Kathleen M. Carley, is a professor in the School of Computer Science in the Institute for Software Research at Carnegie Mellon University. She is has courtesy appointments at Engineering and Public Policy, the Heinz School, and the GSIA. She is the director of the Center for Computational Analysis of Social and Organizational Systems (CASOS), a university wide interdisciplinary center that brings together network analysis, computer science and organization science and has an associated NSF funded training program for Ph.D. students. Kathleen M. Carley's research combines cognitive science, social networks and computer science to address complex social and organizational problems. Her specific research areas are dynamic network analysis, computational social and organization theory, adaptation and evolution, text mining, and the impact of telecommunication technologies and policy on communication, information diffusion, disease contagion and response within and among groups particularly in disaster or crisis situations. She and her lab have developed infrastructure tools for analyzing large scale dynamic networks and various multi-agent simulation systems. She is the founding co-editor with Al Wallace of the journal Computational Organization Theory and has co-edited several books in the computational organizations and dynamic network area.
Computational Intelligence in Cyber Security
Computational Intelligence techniques have proven to be flexible in decision making in dynamic environment. They typically include Fuzzy Logic, Evolutionary Computation, Intelligent Agent Systems, Neural Networks, Cellular Automata, Artificial Immune Systems and other similar computational models. The use of these techniques allowed building efficient and robust decision support modules, providing cross-linking solutions to different cyber security applications.
This tutorial will cover general topics on cyber security, then briefly describe CI techniques and how they are being applied to different security problems such as intrusion detection, spam filtering, etc. The focus of the tutorial will be surveying the state-of-the-art CI-based technologies to cyber security applications.
Dr. Dasgupta’s research interests broadly span the areas of scientific computing, tracking real-world problems through interdisciplinary cooperation. His areas of special interests include Artificial Immune Systems, Genetic Algorithms, Neural Networks, multi-agent systems and their applications. He published more than 150 research papers in book chapters, journals, and international conferences. He authored a book; published two edited volumes and co-edited several conference proceedings over the last 15 years. One of his current researches is applying CI techniques in Network and Internet security, and working on several funded projects. He the founding Director of the Center of Information Assurance at the University of Memphis, a center of Excellence in Information Assurance (CAE/IA), designated by U.S. National Security Agency (NSA) and the Department of Homeland Security (DHS).
The Cubature Kalman filter
Sequential state estimation is a pervasive problem that arises in numerous applications. The problem is difficult to solve when the dynamic system under study is nonlinear. In this tutorial, I will describe a recently discovered and powerful nonlinear filter, which we have named :The Cubature Kalman Filter (CKF)." This discovery is not only important in theoretical but also practical terms.
The discussion begins with a brief review of the Bayesian filter, setting the stage for the rest of the tutorial. I will briefly describe the essence of the cubature rule, and then outline the derivation of the CKF, which provides the bast approximation to the Bayesian filter in the second order sense. This part of the tutorial will finish with a summary of the steps involved in computing the CKF.
In the last part of the tutorial, I will describe practical applications of the CKF:
- aerospace applications;
- supervised training of neural networks; and
- sequential signal detection.
Simon Haykin received his B.Sc. (First-class Honours), Ph.D., and D.Sc., all in Electrical Engineering from the University of Birmingham, England. He is a Fellow of the Royal Society of Canada, and a Fellow of the Institute of Electrical and Electronics Engineers. He is the recipient of the Henry Booker Gold Medal from URSI, 2002, the Honorary Degree of Doctor of Technical Sciences from ETH Zentrum, Zurich, Switzerland, 1999, and many other medals and prizes.
He is a pioneer in adaptive signal-processing with emphasis on applications in radar and communications, an area of research which has occupied much of his professional life. In the mid 1980s, he shifted the thrust of his research effort in the direction of Neural Computation, which was re-emerging at that time. He is currently revisiting the fields of radar and communications from a brand new perspective: Cognitive Radio and Cognitive Radar. These feilds are two important parts of a much wider and multidisciplinary subject: Cognitive Dynamic Systems, research into which has become his passion.